CN104932975A - Test case selection method and device - Google Patents

Abstract

The invention discloses a test case selection method and device. The test case selection method provided by the embodiment comprises the following steps: running a program with each test case in a first test case set, and acquiring coverage information of each test case, corresponding to each sentence in the program, in the first test case set, wherein the first test case set comprises at least one failure test case and a second test case set; adding a part of test cases in the second test case set into a third test case set according to the coverage information of each test case in the first test case set; and locating error sentences in the program through each test case added into the third test case set in order to calculate the anomaly degree of each sentence in the program. Through adoption of the test case selection method provided by the embodiment, the problem of low efficiency of error location since the execution results of all test cases need to be acquired in advance during error location in the prior art is solved.

Description

A kind of test cases selection method and apparatus

Technical field

The present invention relates to computer testing technology, espespecially a kind of test cases selection method and apparatus.

Background technology

Along with the development of computer technology, computer program is as a wherein important developing contents, and carrying out location of mistake to computer program is the major way surveying this program.In order to produce good location of mistake effect, current normally used location of mistake technology needs test use cases that coverage is very high as input, and need coverage information and the execution result of collecting each test case, both are indispensable.

Such as, but under the test scene that some programs lost efficacy, the test case of a large amount of known execution result can not be used for carrying out location of mistake, is in the computer program in performance history.In this case, the quantity of the test case of known execution result is usually considerably less, when these a small amount of known execution result test cases are used for location of mistake, accurately cannot orient the wrong statement of program.At above-mentioned scene Scene, the coverage information of test case can be collected automatically by the method for pitching pile, but the execution result of these test cases is unknown, needs by further judging to obtain execution result; Such as can judge that whether execution result is correct by running all test cases, namely judging whether the execution result of all test cases is consistent with expection; Again such as, the judgement of the execution result of all test cases can also be judged by the Test oracles constructing robotization.

Apparently, prior art when carrying out location of mistake, owing to needing the execution result obtaining all test cases in advance, thus the problem that the efficiency resulting through location of mistake is lower.

Summary of the invention

In order to solve the problems of the technologies described above, the invention provides a kind of test cases selection method and apparatus, to solve prior art when carrying out location of mistake, owing to needing the execution result obtaining all test cases in advance, thus the problem that the efficiency resulting through location of mistake is lower.

First aspect, a kind of test cases selection method provided by the invention, comprising:

Each test case working procedure is concentrated by the first test case, obtaining described first test case concentrates each test case to correspond to the coverage information of every bar statement in program, wherein, described first test use cases comprises at least one failure testing use-case and the second test use cases;

Concentrate the coverage information of each test case according to described first test case, add the described test case of part that described second test case is concentrated to the 3rd test case and concentrate;

The each test case concentrated by the 3rd test case of having added orients the statement of mistake in described program, to calculate the suspicious degree of every bar statement in described program.

In the first possible implementation of first aspect, the initial value of described 3rd test use cases comprises at least one failure testing use-case described, the described coverage information concentrating each test case according to described first test case, by described second test case concentrate the described test case of part add to the 3rd test case concentrate before, also comprise:

Concentrate the coverage information of each test case that the statement of described program is divided into dynamic fundamental block DBB according to described first test case, the statement in described dynamic DBB has identical coverage information;

The coverage information of each test case is concentrated according to current 3rd test case, generate the current largest packet set of described dynamic DBB, largest packet in described current largest packet set comprises one or more described dynamic DBB, and each largest packet in described current largest packet set is concentrated identical test case to cover by described current 3rd test case;

Concentrate the coverage information of each test case according to described first test case, add the described test case of part that described second test case is concentrated to the 3rd test case and concentrate, comprising:

The test case selecting the grouping current group set to be divided into the first grouping and the second grouping is concentrated from current second test case, the described test case selected has different coverage informations to described first grouping and described second grouping, grouping in described current group set comprises one or more described dynamic DBB, and the initial value of grouping set is the largest packet set generated by the initial value of described 3rd test use cases;

The execution result of the test case selected described in acquisition, and the described test case selected is added to described current 3rd test case and concentrate;

When each grouping in described current group set all can not be concentrated test case to divide by described current second test case, judge that whether described current group set is identical with the largest packet set generated by described current 3rd test use cases;

If different, then the grouping in described current group set is carried out merging to form largest packet, thus the coverage information of each test case is concentrated according to described first test case, continue to perform the operation that the described test case of part described second test case concentrated adds the 3rd test use cases to, wherein, being grouped in described current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of described merging;

If identical, then complete the interpolation operation of described 3rd test use cases.

According to the first possible implementation of first aspect, in the implementation that the second is possible, described concentrating from current second test case is selected the grouping current group set to be divided into the first grouping and the second test case of dividing into groups, and comprising:

The test case selecting described grouping to be divided into uniformly described first grouping and described second grouping is concentrated from described current second test case.

The implementation possible according to the first or the second of first aspect, in the implementation that the third is possible, concentrate the test case of selection to meet the following conditions from described current second test case:

$\arg \underset{t\∈T_g}{\max }\{\underset{i=1}{\overset{\left|G\left(T_r\right)\right|}{\mathrm{\Σ}}}S\left(g_i\right)*\mathrm{Split}(t,g_i)\},$ Wherein, Split (t, g _i)=min (| g _i1|, | g _i2|)

Wherein, S (g _i) represent the grouping g covered by the described test case t selected _iin comprise the degree of mistake statement, | G (T _r) | represent according to described current 3rd test use cases T _rthe quantity of the grouping formed, Split (t, g _i) the test case t that selected described in expression is by described grouping g _ibe divided into described first grouping g _i1with described second grouping g _i2in the quantity of a less dynamic DBB comprised.

Second aspect, a kind of test cases selection device provided by the invention, comprising:

Acquisition module, for concentrating each test case working procedure by the first test case, obtaining described first test case concentrates each test case to correspond to the coverage information of every bar statement in program, wherein, described first test use cases comprises at least one failure testing use-case and the second test use cases;

Add module, the first test case for obtaining according to described acquisition module concentrates the coverage information of each test case, adds the described test case of part that described second test case is concentrated to the 3rd test case and concentrates;

Location of mistake module, each test case that the 3rd test case for being added by described interpolation module is concentrated orients the statement of mistake in described program, to calculate the suspicious degree of every bar statement in described program.

In the first possible implementation of second aspect, the initial value of described 3rd test use cases comprises at least one failure testing use-case described, described device also comprises: the first generation module, for concentrating the coverage information of each test case in described interpolation module according to the first test case that described acquisition module obtains, by described second test case concentrate the described test case of part add to the 3rd test case concentrate before, concentrate the coverage information of each test case that the statement of described program is divided into dynamic fundamental block DBB according to the first test case that described acquisition module obtains, statement in described dynamic DBB has identical coverage information,

Second generation module, for concentrating the coverage information of each test case according to current 3rd test case, generate the current largest packet set of described dynamic DBB, largest packet in described current largest packet set comprises one or more described dynamic DBB, and each largest packet in described current largest packet set is concentrated identical test case to cover by described current 3rd test case;

Then add module, comprise: selection unit, for concentrating the test case selecting the grouping in current group set to be divided into the first grouping and the second grouping from current second test case, the described test case selected has different coverage informations to described first grouping and described second grouping, grouping in described current group set comprises one or more described dynamic DBB, and the initial value of grouping set is the largest packet set generated by the initial value of described 3rd test use cases;

Adding device, for obtaining the execution result of the test case that described selection unit has been selected, and adds described current 3rd test case to and concentrates by the described test case selected;

Judging unit, for when each grouping in described current group set all can not be concentrated test case to divide by described current second test case, judge that whether described current group set is identical with the largest packet set generated by described current 3rd test use cases;

Then described adding device, time also for being different in the judged result of described judging unit, grouping in described current group set is carried out merging to form largest packet, thus the coverage information of each test case is concentrated according to described first test case, continue to perform the operation that the described test case of part described second test case concentrated adds the 3rd test use cases to, wherein, being grouped in described current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of described merging; Or,

Described adding device, also for when the judged result of described judging unit is identical, completes the interpolation operation of described 3rd test use cases.

According to the first possible implementation of second aspect, in the implementation that the second is possible, described selection unit, specifically for concentrating the test case selecting described grouping to be divided into uniformly described first grouping and described second grouping from described current second test case.

The implementation possible according to the first or the second of second aspect, in the implementation that the third is possible, described selection unit concentrates the test case of selection to meet the following conditions from described current second test case:

$\arg \underset{t\∈T_g}{\max }\{\underset{i=1}{\overset{\left|G\left(T_r\right)\right|}{\mathrm{\Σ}}}S\left(g_i\right)*\mathrm{Split}(t,g_i)\},$ Wherein, Split (t, g _i)=min (| g _i1|, | g _i2|)

Wherein, S (g _i) represent the grouping g covered by the described test case t selected _iin comprise the degree of mistake statement, | G (T _r) | represent according to described current 3rd test use cases T _rthe quantity of the grouping formed, Split (t, g _i) the test case t that selected described in expression is by described grouping g _ibe divided into described first grouping g _i1with described second grouping g _i2in the quantity of a less dynamic DBB comprised.

Test cases selection method and apparatus provided by the invention, each test case working procedure is concentrated by the first test case, obtaining this first test case concentrates each test case to correspond to the coverage information of program statement, this first test use cases comprises at least one failure testing use-case and the second test use cases, thus concentrate a selection portion point test case to add the 3rd test case to from the second test case to concentrate according to the coverage information obtained, and then by each test case that the 3rd test case of having added is concentrated, location of mistake is carried out to program; The method that the present embodiment provides carries out location of mistake by the test case selected, achieve, by less test case, location of mistake is carried out to the statement in program, to solve prior art when carrying out location of mistake, owing to needing the execution result obtaining all test cases in advance, thus the problem that the efficiency resulting through location of mistake is lower, correspondingly improve the work efficiency of debugged program.

Accompanying drawing explanation

Accompanying drawing is used to provide the further understanding to technical solution of the present invention, and forms a part for instructions, is used from and explains technical scheme of the present invention, do not form the restriction to technical solution of the present invention with the embodiment one of the application.

The process flow diagram of a kind of test cases selection method that Fig. 1 provides for the embodiment of the present invention;

The process flow diagram of the another kind of test cases selection method that Fig. 2 provides for the embodiment of the present invention;

The structural representation of a kind of test cases selection device that Fig. 3 provides for the embodiment of the present invention;

The structural representation of the another kind of test cases selection device that Fig. 4 provides for the embodiment of the present invention.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly understand, hereinafter will be described in detail to embodiments of the invention by reference to the accompanying drawings.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combination in any mutually.

Can perform in the computer system of such as one group of computer executable instructions in the step shown in the process flow diagram of accompanying drawing.Further, although show logical order in flow charts, in some cases, can be different from the step shown or described by order execution herein.

The process flow diagram of a kind of test cases selection method that Fig. 1 provides for the embodiment of the present invention.The test cases selection method that the present embodiment provides is applicable to the test case selected from the test case of a large amount of unknown result for carrying out location of mistake, the method can be performed by test cases selection device, this device realizes with the method for hardware and software usually, can be integrated in the processor of terminal device, such as be integrated in the storage unit of processor chips, call execution for processor.As shown in Figure 1, the method for the present embodiment can comprise:

S110, each test case working procedure is concentrated by the first test case, obtaining the first test case concentrates each test case to correspond to the coverage information of program statement, and this first test use cases comprises at least one failure testing use-case and the second test use cases.

In the present embodiment, the test case that second test case is concentrated is the test case of unknown result, the test case of those unknown results can be generated by Test cases technology instrument, have at present and can generate Test cases technology instrument that is complete, the reasonably robotization of test input, by this Test cases technology instrument, the test input meeting specific covering demand in a large number can be generated, as branch covers.In specific implementation, when programmer carries out location of mistake to a program developed, first a large amount of test inputs can be generated by existing Test cases technology instrument, to obtain the test case with high coverage, namely all tested use-case of the executable statement in program covers, and generally includes the test case of a large amount of unknown result; In addition, it is failed test case that the debugging due to program originates at least one result usually, hereinafter referred to as failure testing use-case, thus at source code P and at least one failure testing use-case above-mentioned of initial time known procedure.Therefore, each test case concentrated by running at least one failure testing use-case above-mentioned and the second test case can get the coverage information that the first test case concentrates the corresponding program statement of each test case.

For example, in the present embodiment, the test use cases of the unknown execution result generated by Test cases technology instrument is labeled as T _g, this T _gbe exactly the second test use cases, and suppose program P is made up of m statement, can be expressed as P={s ₁, s ₂..., s _m.The first test use cases operated on program P is expressed as T=T _g∪ T _s, wherein, T _sit is the set of at least one failure testing use-case above-mentioned; Using the test case in the first test use cases T as input executive routine P, by the coverage information of the method record statement of pitching pile in a program.Particularly, after the every bar statement by each test case working procedure in the first test use cases T, the coverage information that can collect is designated as MS, this MS can be expressed as n × m and tie up matrix, wherein, n is the quantity of test case in the first test use cases T, each element c of matrix _ijrepresent by test case t _iin the process of working procedure, statement s _jwhether be capped, if capped, c _ijvalue be 1, otherwise c _ijvalue be 0.Now, get the coverage information that each test case in the first test use cases T corresponds to every bar statement in program P, but, above-mentioned second test use cases T _gin the execution result of each test case be unknown.

S120, concentrates the coverage information of each test case according to the first test case, add the partial test use-case that this second test case is concentrated to the 3rd test case and concentrate.

Adopt in prior art and judge above-mentioned second test use cases T one by one _gin the mode of execution result of each test case, whole test cases that automatic test cases Core Generator generates are applied to location of mistake technology, but, judge that execution result corresponding to a large amount of test case can produce a large amount of manpower costs or system overhead, not only can increase the workload of programmer, also can increase the debugging burden in trace routine process, thus reduce the work efficiency of mistake in locator meams trace routine.

In the present embodiment, by the above-mentioned coverage information MS obtained, from the second test use cases T _gmiddle selection part test case is added the 3rd test case to and is concentrated, and is added into the test case that the 3rd test case concentrates and is the test case needing judged result, the 3rd test use cases is labeled as T _r, for representing the test use cases of known results.It should be noted that, before performing the operation of the interpolation in the present embodiment, this T _rinitial value be T _s, due to T _sin the result of each test case known, be failure, therefore do not starting to carry out from the second test use cases T _gbefore the work of middle selection, T _rin there is T _s.

S130, each test case concentrated by described 3rd test case of having added orients the statement of mistake in program, to calculate the suspicious degree of every bar statement in described program.

In the present embodiment, according to the 3rd test use cases T added _rin each test case location of mistake is carried out to program, be about to the test case of a small amount of known results selected for location of mistake, this 3rd test use cases T added _rcomprise above-mentioned T _swith from T _gthe test case of middle selection, and this T _rthe quantity of middle test case is less than the quantity of test case in T, thus passes through T _rthe suspicious degree of the test case calculation procedure statement of middle known results, such as, can generate the seniority among brothers and sisters of suspicious statement, thus programmer can investigate statement successively according to the height of the suspicious degree of statement.Apparently, in the present embodiment, judgement is only needed to add the 3rd test use cases T _rin the execution result of each test case, the mode of this judgement can be such as, judges that whether the operation result of test case to be added is consistent with expected results, if unanimously, then this test case is labeled as correct test case, otherwise is labeled as failure testing use-case; It should be noted that, the behavior of above-mentioned judgement and mark can be performed by computer program, from row labels after also can being judged according to expected results and operation result by programmer, only needs acquisition to add the 3rd test use cases T in the present embodiment _rin the execution result of test case, considerably reduce the time and efforts judging that test case execution result spends.

In specific implementation, the method that this present embodiment provides can be applied in the location of mistake technology based on coverage information, and the Typical Representative of this class location of mistake technology has Tarantula, Ochiai etc.When after a certain location of mistake technology of employing, with the 3rd test use cases T selected _ras input, just can obtain the seniority among brothers and sisters of each executable statement in program, programmer can rank according to this and carry out debugging activity.

The test cases selection method that the present embodiment provides, each test case working procedure is concentrated by the first test case, obtaining this first test case concentrates each test case to correspond to the coverage information of program statement, this first test use cases comprises at least one failure testing use-case and the second test use cases, thus concentrate a selection portion point test case to add the 3rd test case to from the second test case to concentrate according to the coverage information obtained, and then by each test case that the 3rd test case of having added is concentrated, location of mistake is carried out to program; The method that the present embodiment provides carries out location of mistake by the test case selected, achieve, by less test case, location of mistake is carried out to the statement in program, to solve prior art when carrying out location of mistake, owing to needing the execution result obtaining all test cases in advance, thus the problem that the efficiency resulting through location of mistake is lower, correspondingly improve the work efficiency of debugged program.

The process flow diagram of the another kind of test cases selection method that Fig. 2 provides for the embodiment of the present invention.On above-mentioned basis embodiment illustrated in fig. 1, also comprise before S120:

S111, according to the first test case concentrate the coverage information of each test case the statement of program is divided into dynamic fundamental block (Dynamic Basic Block, referred to as: DBB), the statement in this dynamic DBB has identical coverage information.

Dynamic DBB in the present embodiment divides according to the coverage information of each test case in the first test use cases T, and namely this dynamic DBB marks off based on the coverage information of test case, is different from DBB of the prior art, divides based on program structure.Owing to there is the statement that some can not be made a distinction by any test case in a program, for the test use cases collecting statement coverage information, some statements in program are always performed by same test case, namely these statements have identical coverage information, represent the above-mentioned statement with identical coverage information in the present embodiment with dynamic DBB; Namely dynamic DBB is defined as: suppose there is program P and test use cases T _o, when some the statement set in program P meet two following conditions, these statement set are a dynamic DBB, and (1), these statements are all by T _oin identical test case cover; (2) if k is a statement in dynamic DBB, and h is not the statement in set DBB, so T _oin at least there is a test case t _x, passing through this test case t _xone in statement k and h is only covered when working procedure.

Below by way of the dynamic DBB that an instantiation proposes the present embodiment, and the grouping of follow-up proposition and the concept of largest packet are explained.As shown in table 1 below, be the coverage information corresponding to every bar statement in pow program (title of program) by Fig. 1 or acquisition embodiment illustrated in fig. 2.

Table 1

As shown in table 1, t ₁for the test case that result is failed, be T _sin element, the T of example shown in table 1 _sin only by element, i.e. a t ₁, t ₂~ t ₅be the second test use cases T _gin element, this pow program is specially an exponent arithmetic program, for calculating the y power of x, the right side of table 1 lists the first test use cases T operated in pow program, have 5 test cases, under input corresponding to each test case is presented at this test case title, the situation that each test case corresponds to the coverage information of every bar statement in program represents by "●" and " blank " respectively, if test case covers corresponding statement in program process, then "●" on corresponding table position mark, otherwise corresponding table position is blank, the execution result of test case is presented at last column of form, symbol P and F represents that test case execution result is for correct and failure respectively, initial time, only obtain test case t ₁execution result be unsuccessfully, other test case is the test case of unknown result.In the present embodiment, according to the coverage information shown in table 1, the statement in pow program can be divided into dynamic DBB, as shown in table 2 below, for by the dynamic DBB that the statement of the program of pow shown in table 1 is marked off embodiment illustrated in fig. 2.

Table 2

Dynamic DBB	t 1	t 2	t 3	t 4	t 5
						b 1{1,2,3,6,9,11}	●	●	●	●	●
b 2{4,10}	●	●	●
						b 3{5}				●	●
b 4{7,8}	●		●	●

As shown in table 2, all statements in each dynamic DBB have identical coverage information, incorporated by reference to table 1, and such as b ₂in statement 4 and 10, all by t ₁, t ₂and t ₃covered, and not by t ₄and t ₅covered.

It should be noted that, because DBB of the prior art is difficult to distinguish the situation that is performed of statement in program, namely cannot divide DBB by the coverage information of test case, DBB of the prior art only can represent the statement sequence that in program, one section of order performs, wherein only have an entrance and an outlet, entrance is exactly the first statement of DBB, outlet is exactly the last item statement of DBB, namely the structure of program can only be embodied, represent the coverage information of test case according to this DBB, can deviation be produced.Therefore, propose the concept of dynamic DBB in the present embodiment, represent the coverage information of test case using this dynamic DBB as base unit.

S112, the coverage information of each test case is concentrated according to current 3rd test case, generate the current largest packet set of dynamic DBB, largest packet in this current largest packet set comprises one or more dynamic DBB, and each largest packet in this current largest packet set is concentrated identical test case to cover by current 3rd test case.

Divided in the actual conditions of dynamic DBB by coverage information, if two statements are covered by the correct test case of the failure testing use-case of equivalent amount and equivalent amount, be so difficult to by above-mentioned coverage information, these two statements be made a distinction; For above-mentioned situation, in order to portray the division ability of coverage information to program statement of test case further, the present embodiment describes dynamic DBB at the 3rd test use cases T by the concept of grouping and largest packet _ron indistinguishability.

Particularly, define dynamic DBB and gather for grouping, represent with g, the following condition of grouping g demand fulfillment: (1), grouping g are DBB (T _r) a nonvoid subset, DBB (T _r) for passing through T _rthe DBB set formed; (2), remember that f (b) represents the 3rd test use cases T _rin cover the quantity of the failure testing use-case of dynamic DBB, p (b) represents the 3rd test use cases T _rin cover the quantity of the correct test case of dynamic DBB, then such as, for any two dynamic DBB, u and v in grouping g, meet f (u)=f (v) and p (u)=p (v).

The set of above-mentioned grouping is with G (T _r) represent, if this G (T _r) in one grouping g meet the following conditions, then g is this grouping set G (T _r) in largest packet; Condition is: for G (T _r) any one grouping g', if g' ≠ g, then for grouping g in any one dynamic DBB, be such as u, and any one the dynamic DBB in g', be such as v, demand fulfillment: f (u) ≠ f (v) or p (u) ≠ p (v); And the set formed with largest packet, for largest packet set, is designated as DG (T _r).

That is, for grouping g ₁and g ₂, covered by the correct test case of the failure testing use-case of equivalent amount and equivalent amount, this g ₁and g ₂can be two independently to divide into groups, but also can be a largest packet (g ₁, g ₂).

After obtained above-mentioned dynamic DBB by coverage information, can generate and pass through T _rthe largest packet set formed, at initial time, T _rfor T _s, be explained for the program of pow shown in table 1 equally, T _rin only have the test case t of a known execution result ₁, i.e. T _sin only comprise element t ₁, t ₁the dynamic DBB covered comprises: b ₁, b ₂and b ₄, i.e. b ₁, b ₂and b ₄form a largest packet, meanwhile, b ₃also form a largest packet, therefore, pass through T _r={ t ₁the largest packet set that formed is specially: DG (T _r)=G (T _r)={ (b ₁, b ₂, b ₄), (b ₃).

In the present embodiment, T is passed through in acquisition _r={ t ₁after the largest packet set that formed, carry out the second test use cases T _gin partial test use-case add the 3rd test use cases T to _rin operation, namely S120 comprises:

S121, the test case selecting the grouping current group set to be divided into the first grouping and the second grouping is concentrated from current second test case, this test case selected has different coverage informations to described first grouping and described second grouping, grouping in this current grouping set comprises one or more dynamic DBB, and the initial value of this current grouping set is the largest packet set generated by the initial value of the 3rd test use cases.

The present embodiment, in specific implementation, divides the grouping in grouping set first, at initial time, is actually passing through T _r=T _slargest packet in the largest packet set formed divides, if the second test use cases T _gin a test case t cover only grouping g in part dynamic DBB, according to grouping definition, grouping g is divided into two groupings by test case t, such as, be the first grouping g ₁with the second grouping g ₂.

Alternatively, from the second test use cases T _gwhen middle selection can divide the test case of grouping g, preferentially evenly can divide the test case of grouping g, if because the first grouping g produced after test case division grouping g ₁with the second grouping g ₂size identical or relatively, so the first grouping g ₁with the second grouping g ₂when dividing, then the test case used is less again.

In specific implementation, set current grouping set G (T _r) grouping that comprises can be expressed as { g ₁, g ₂... g _u, when consideration is from the second test use cases T _gmiddle selection test case t adds the 3rd test use cases T _rtime, t can meet the following conditions:

$\arg \underset{t\∈T_g}{\max }\{\underset{i=1}{\overset{\left|G\left(T_r\right)\right|}{\mathrm{\Σ}}}S\left(g_i\right)*\mathrm{Split}(t,g_i)\},$ Wherein, Split (t, g _i)=min (| g _i1|, | g _i2|) (1)

S (g in above-mentioned (1) formula _i) represent the grouping g covered by the test case t selected _iin comprise the degree of mistake statement, | G (T _r) | represent according to current 3rd test use cases T _rthe quantity of the grouping formed.

S (g in above-mentioned (1) formula _i) concrete account form be:

$S\left(g_i\right)=a_{\mathrm{ef}}\left(g_i\right)-\frac{a_{\mathrm{ep}}\left(g_i\right)}{a_{\mathrm{ep}}\left(g_i\right)+a_{\mathrm{np}}\left(g_i\right)+1}---\left(2\right)$

Wherein, a in above-mentioned (2) formula _ep(g _i) represent g _iby T _rin the quantity that covers of correct test case, a _ef(g _i) represent g _iby T _rthe quantity that middle failure testing use-case covers, a _np(g _i) represent g _inot by T _rin the quantity that covers of correct test case.

It should be noted that, Split (t, g in (1) formula _i) represent that test case t is to grouping g _idivide the first grouping g produced _i1with the second grouping g _i2in the less quantity comprising dynamic DBB; If test case t is not to g _iproduce and divide, then Split (t, g _i) be 0, and Split (t, g _i) paying the utmost attention to can to g _iproduce the test case t evenly divided as far as possible, produce two after namely dividing and comprise the first close as far as possible grouping g of dynamic DBB quantity _i1with the second grouping g _i2.

S122, obtains the execution result of test case selected, and the test case selected is added to current 3rd test case and concentrate.

In the present embodiment, the test case selected in S121 is usually enough to be divided the potential grouping comprising mistake statement, adds the 3rd test use cases T to for selection _rin test case, for follow-up carry out location of mistake time need the test case that uses, therefore, need the execution result obtaining the test case selected, thus meet the condition of location of mistake.

Further, if from the second test use cases T _gmiddle selection some add the 3rd test use cases T _rin test case, now, each grouping in current group set all can not be concentrated T by current second test case _gtest case divides, but may there is above-mentioned situation, some groupings namely in grouping set may be covered by the correct test case of the failure testing use-case of equal number and equal number, dynamic DBB in these groupings can not be distinguished, and the second current test use cases T _gin still may exist some test cases cover only in these grouping sets a part grouping, namely may need continue thus current second test use cases T _gmiddle selection is for carrying out the test case of location of mistake, then the method that provides of the present embodiment, also comprises in S120:

S123, when each grouping in current group set all can not be concentrated test case to divide by current second test case, judges that whether current group set is identical with the largest packet set generated by current 3rd test use cases; If different, then perform S124, if identical, then perform S125.It should be noted that, performing the largest packet set being used for contrasting with current group set when judging in S123 is, the largest packet set that current 3rd test use cases obtained by S112 is generated, and the 3rd test use cases is now the 3rd test use cases added after performing S122.

S124, is undertaken merging to form largest packet by the grouping in current group set, wherein, and being grouped in current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of merging.After merging, concentrate the coverage information of each test case according to the first test case, continue to perform the operation that the partial test use-case the second test case concentrated adds the 3rd test use cases to, after namely performing S124, return S121 and continue to perform.

S125, completes the interpolation operation of the 3rd test use cases.

In the present embodiment, when each grouping in current group set all can not be concentrated test case to divide by current second test case, need further by current the 3rd test use cases T selected _rwhether the largest packet set generated is identical with the grouping set now after repeatedly dividing, if different, then the second test use cases T is described _gmiddle may exist some test cases cover only in current group set part grouping, therefore need consider these test cases selection are added the 3rd test use cases T _rin, if identical, then illustrate that current group set can not be divided again, i.e. current 3rd test use cases T _rfor finally needing the 3rd test use cases selected.

Equally for the dynamic DBB to pow procedure division shown in pow program shown in above-mentioned table 1 and above-mentioned table 2, the dividing mode of dividing into groups in the present embodiment is described, is passing through initial T _r={ t ₁the largest packet set DG (T that formed _r)=G (T _r)={ (b ₁, b ₂, b ₄), (b ₃) after, starting selection needs to insert the 3rd test use cases T _rin test case, such as, first select test case t ₂, by judging test case t ₂execution result be labeled as failure testing use-case, use test case t ₂test, b ₄coverage information different, therefore b ₄from largest packet (b ₁, b ₂, b ₄) in divide separated, now, T _r={ t ₁, t ₂, G (T _r)={ (b ₁, b ₂), (b ₃), (b ₄); Subsequently, have selected test case t in this example ₄, continue test case t ₄execution result marked after, T _r={ t ₁, t ₂, t ₄, now G (T _r)={ (b ₁), (b ₂), (b ₃), (b ₄).Due to the second test use cases T _gin arbitrary test case all can not again to G (T _r) in grouping divide, and | G (T _r) |=| DG (T _r) |, thus the selection course of test case completes, and namely final the 3rd test use cases selected is: T _r={ t ₁, t ₂, t ₄.It should be noted that, select test case t ₄shi Ruo selects test case t ₅identical effect can be realized.

Obtaining the 3rd test use cases T selected _rafter, the 3rd test use cases T that this can have been selected _rfor location of mistake, adopt the suspicious degree of location of mistake technology Ochiai computing statement, the suspicious degree rank of each statement can be obtained, as shown in table 3, be the suspicious angle value of statement and the seniority among brothers and sisters thereof of the pow program by generation embodiment illustrated in fig. 2; Can see, statement s ₄and s ₁₀suspicious degree seniority among brothers and sisters be 2, if there is several statement come to the same thing during seniority among brothers and sisters, all get the worst result, due to s ₄and s ₁₀suspicious degree identical, illustrate and at most only need check that two statements just can find error code, and also need to check the execution result of 2 test cases in test cases selection process.

Table 3

Statement number	s 1	s 2	s 3	s 4	s 5	s 6	s 7	s 8	s 9	s 10	s 11
												Suspicious angle value	0.50	0.50	0.50	1.00	0.00	0.50	0.33	0.33	0.50	1.00	0.50
Seniority among brothers and sisters	8	8	8	2	11	8	10	10	8	2	8

It should be noted that, during the largest packet set of pow Program Generating shown in above-mentioned table 1 is divided, after completing division, the result divided grouping set is that each dynamic DBB is a grouping, and cannot divide packet combining, that is, T is passed through _r={ t ₁, t ₂, t ₄the largest packet set D (T that formed _r)=G (T _r)={ (b ₁), (b ₂), (b ₃), (b ₄), therefore, T _r={ t ₁, t ₂, t ₄be final the 3rd test case set selected, select test case without the need to merging rear continuation to the grouping in grouping set again.

Below by way of the implementation that one section of concrete code description is embodiment illustrated in fig. 2, as shown in table 4, be algorithmic procedure embodiment illustrated in fig. 2.

Table 4

The input of algorithm is failure testing set of uses case T _swith the second test use cases T of a large amount of execution result the unknown _gand statement coverage information MS, output is the 3rd test use cases T for location of mistake _r.Algorithm mainly comprises step main as follows:

(1), algorithm 1 ~ 4 row is first by initial failure testing use-case T _sadd T _r, simultaneously according to the first test use cases T=T _g∪ T _scoverage information generate dynamic DBB, i.e. DBB (T), the transverse direction in corresponding table 2, the information that the tested use-case of dynamic DBB covers, and in units of dynamic DBB, represent test case, the longitudinal direction in corresponding table 2, the information that test case is covered by dynamic DBB.

(2), first algorithm 6 row generates initial second test use cases T _r=T _stime largest packet set DG (T _r), then with this largest packet set as current grouping set G (T _r).

(3), algorithm 7 ~ 16 row iteration is from T _gmiddle selection test case adds T _r; Particularly, 11 ~ 12 row judge whether the test case of current selection divides the grouping in current group set, if do not produce division, then stops the selection course of test case, otherwise after the test case mark execution result selected, this test case is added T _r, and repeat again to select new test case.

(4), above-mentionedly to illustrate, as grouping set G (T _r) in each grouping when can not be divided again, due to G (T _r) in some groupings may be covered by the correct test case of the failure testing use-case of equal number and equal number, the dynamic DBB in these groupings can not be distinguished, and current T _gstill may there are some test cases and cover only part grouping in these grouping sets, next step of thus algorithm also considers these test cases selection to join T _rin, i.e. 17 ~ 19 row of algorithm, first can according to current T _robtain current largest packet set, thus judge that whether current group set is identical with the current largest packet set of acquisition, if not identical, grouping in current group set can be merged, packet combining here refer to by grouping set by the packet combining of the failure testing use-case covering of the correct of equal number and equal number in same grouping to form largest packet; If be judged as identical, then illustrate that current grouping set can not divide by tested use-case again, algorithm convergence, returns the 3rd test use cases T selected _r.It should be noted that, algorithm judge not identical after executive mode again largest packet set is divided again as skipping to the 6th row.

The above embodiment of the present invention is by proposing based on the dynamic DBB of coverage information and grouping and the dividing mode of largest packet, each statement in program can be divided with dynamic DBB and grouping, and by dividing the statement by being difficult in prior art distinguish further the division of grouping, to realize by a small amount of test case, namely can carry out efficient location of mistake effect, the accuracy of location of mistake can be ensured simultaneously.

The structural representation of a kind of test cases selection device that Fig. 3 provides for the embodiment of the present invention.The test cases selection device that the present embodiment provides is applicable to the test case selected from the test case of a large amount of unknown result for carrying out location of mistake, this device realizes in the mode of hardware and software usually, can be integrated in the processor of terminal device, such as be integrated in the storage unit of processor chips, call execution for processor.As shown in Figure 3, the test cases selection device of the present embodiment specifically comprises: acquisition module 11, interpolation module 12 and location of mistake module 13.

Wherein, acquisition module 11, for concentrating each test case working procedure by the first test case, obtaining the first test case concentrates each test case to correspond to the coverage information of every bar statement in program, wherein, the first test use cases comprises at least one failure testing use-case and the second test use cases.

Add module 12, the first test case for obtaining according to acquisition module 11 concentrates the coverage information of each test case, adds the partial test use-case that the second test case is concentrated to the 3rd test case and concentrates.

In the present embodiment, the second test case concentrates each test case comprised to be the test case of unknown result, and the initial value of the 3rd test use cases equally only comprises at least one failure testing use-case above-mentioned.

Location of mistake module 13, each test case that the 3rd test case for adding by adding module 12 is concentrated orients the statement of mistake in program, to calculate the suspicious degree of every bar statement in program.

The test cases selection device that the embodiment of the present invention provides is for performing the present invention's test cases selection method provided embodiment illustrated in fig. 1, and possess corresponding functional module, it realizes principle and technique effect is similar, repeats no more herein.

The structural representation of the another kind of test cases selection device that Fig. 4 provides for the embodiment of the present invention, on above-mentioned basis embodiment illustrated in fig. 3, the device that the present embodiment provides also comprises: the first generation module 14, for concentrating the coverage information of each test case according to the first test case that acquisition module 11 obtains in interpolation module 12, by the second test case concentrate partial test use-case add to the 3rd test case concentrate before, concentrate the coverage information of each test case that the statement of program is divided into dynamic DBB according to the first test case that acquisition module 11 obtains, statement in dynamic DBB has identical coverage information, second generation module 15, for concentrating the coverage information of each test case according to current 3rd test case, generate the current largest packet set of dynamic DBB, largest packet in current largest packet set comprises one or more dynamic DBB, and each largest packet in current largest packet set is concentrated identical test case to cover by current 3rd test case.

Correspondingly, interpolation module 12 in the present embodiment can comprise: selection unit 16, for concentrating the test case selecting the grouping in current group set to be divided into the first grouping and the second grouping from current second test case, this test case selected has different coverage informations to the first grouping and the second grouping, and the grouping in current group set comprises one or more dynamic DBB; Adding device 17, for obtaining the execution result of the test case that selection unit 16 has been selected, and adds current 3rd test case to and concentrates by the test case selected.It should be noted that, the present embodiment is when dividing the grouping in current group set first, and the object of division, current group set namely to be divided is the largest packet set generated by the initial value of the 3rd test use cases.

Alternatively, the selection unit 16 in the present embodiment, specifically for concentrating the test case selecting the grouping in grouping set to be divided into uniformly the first grouping and the second grouping from current second test case.Such as by above-mentioned (1) formula and (2) formula, grouping is divided equally.

Further, the present embodiment is after dividing the grouping in grouping set, some groupings existed equally in grouping set may be covered by the correct test case of the failure testing use-case of equal number and equal number, now, dynamic DBB in these groupings can not be distinguished, but, the second current test use cases still may exist some test cases cover only in these grouping sets a part grouping, namely may need to continue thus current second test case is concentrated and selected test case for carrying out location of mistake, therefore in the device that the present embodiment provides, add module 12 also to comprise: judging unit 18, when all test case can not be concentrated to divide by current second test case for each grouping in current group set, judge that whether current group set is identical with the largest packet set generated by current 3rd test use cases, correspondingly, adding device 17, time also for being different in the judged result of judging unit 18, grouping in current group set is carried out merging to form largest packet, thus the coverage information of each test case is concentrated according to the first test case, continue to perform the operation that the partial test use-case the second test case concentrated adds the 3rd test use cases to, wherein, being grouped in current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of merging, or adding device 17, also for when the judged result of judging unit 18 is identical, completes the interpolation operation of the 3rd test use cases.

The test cases selection device that the embodiment of the present invention provides is for performing the present invention's test cases selection method provided embodiment illustrated in fig. 2, possesses corresponding functional module, dynamic DBB is divided in the present embodiment, generate the largest packet set corresponding to the 3rd test use cases, grouping set is divided, and carry out judging to select further the beneficial effect of the concrete mode of test case and realization all same as the previously described embodiments after dividing, therefore do not repeat them here.

Although the embodiment disclosed by the present invention is as above, the embodiment that described content only adopts for ease of understanding the present invention, and be not used to limit the present invention.Those of skill in the art belonging to any the present invention; under the prerequisite not departing from the spirit and scope disclosed by the present invention; any amendment and change can be carried out in the form implemented and details; but scope of patent protection of the present invention, the scope that still must define with appending claims is as the criterion.

Claims (8)

1. a test cases selection method, is characterized in that, comprising:

Each test case working procedure is concentrated by the first test case, obtaining described first test case concentrates each test case to correspond to the coverage information of every bar statement in program, wherein, described first test use cases comprises at least one failure testing use-case and the second test use cases;

Concentrate the coverage information of each test case according to described first test case, add the described test case of part that described second test case is concentrated to the 3rd test case and concentrate;

The each test case concentrated by the 3rd test case of having added orients the statement of mistake in described program, to calculate the suspicious degree of every bar statement in described program.

2. test cases selection method according to claim 1, it is characterized in that, the initial value of described 3rd test use cases comprises at least one failure testing use-case described, the described coverage information concentrating each test case according to described first test case, by described second test case concentrate the described test case of part add to the 3rd test case concentrate before, also comprise:

Concentrate the coverage information of each test case that the statement of described program is divided into dynamic fundamental block DBB according to described first test case, the statement in described dynamic DBB has identical coverage information;

The coverage information of each test case is concentrated according to current 3rd test case, generate the current largest packet set of described dynamic DBB, largest packet in described current largest packet set comprises one or more described dynamic DBB, and each largest packet in described current largest packet set is concentrated identical test case to cover by described current 3rd test case;

Concentrate the coverage information of each test case according to described first test case, add the described test case of part that described second test case is concentrated to the 3rd test case and concentrate, comprising:

The test case selecting the grouping current group set to be divided into the first grouping and the second grouping is concentrated from current second test case, the described test case selected has different coverage informations to described first grouping and described second grouping, grouping in described current group set comprises one or more described dynamic DBB, and the initial value of grouping set is the largest packet set generated by the initial value of described 3rd test use cases;

The execution result of the test case selected described in acquisition, and the described test case selected is added to described current 3rd test case and concentrate;

When each grouping in described current group set all can not be concentrated test case to divide by described current second test case, judge that whether described current group set is identical with the largest packet set generated by described current 3rd test use cases;

If different, then the grouping in described current group set is carried out merging to form largest packet, thus the coverage information of each test case is concentrated according to described first test case, continue to perform the operation that the described test case of part described second test case concentrated adds the 3rd test use cases to, wherein, being grouped in described current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of described merging;

If identical, then complete the interpolation operation of described 3rd test use cases.

3. test cases selection method according to claim 2, is characterized in that, described concentrating from current second test case is selected the grouping current group set to be divided into the first grouping and the second test case of dividing into groups, and comprising:

The test case selecting described grouping to be divided into uniformly described first grouping and described second grouping is concentrated from described current second test case.

4. the test cases selection method according to Claims 2 or 3, is characterized in that, concentrates the test case of selection to meet the following conditions from described current second test case:

$a\mathrm{rg}\underset{t\∈T_g}{\max }\{\underset{i=1}{\overset{\left|G\left(T_r\right)\right|}{\mathrm{\Σ}}}S\left(g_i\right)*\mathrm{Split}(t,g_i)\},$ Wherein, Split (t, g _i)=min (| g _i1|, | g _i2|)

Wherein, S (g _i) represent the grouping g covered by the described test case t selected _iin comprise the degree of mistake statement, | G (T _r) | represent according to described current 3rd test use cases T _rthe quantity of the grouping formed, Split (t, g _i) the test case t that selected described in expression is by described grouping g _ibe divided into described first grouping g _i1with described second grouping g _i2in the quantity of a less dynamic DBB comprised.

5. a test cases selection device, is characterized in that, comprising:

Acquisition module, for concentrating each test case working procedure by the first test case, obtaining described first test case concentrates each test case to correspond to the coverage information of every bar statement in program, wherein, described first test use cases comprises at least one failure testing use-case and the second test use cases;

Add module, the first test case for obtaining according to described acquisition module concentrates the coverage information of each test case, adds the described test case of part that described second test case is concentrated to the 3rd test case and concentrates;

Location of mistake module, each test case that the 3rd test case for being added by described interpolation module is concentrated orients the statement of mistake in described program, to calculate the suspicious degree of every bar statement in described program.

6. test cases selection device according to claim 5, it is characterized in that, the initial value of described 3rd test use cases comprises at least one failure testing use-case described, described device also comprises: the first generation module, for concentrating the coverage information of each test case in described interpolation module according to the first test case that described acquisition module obtains, by described second test case concentrate the described test case of part add to the 3rd test case concentrate before, concentrate the coverage information of each test case that the statement of described program is divided into dynamic fundamental block DBB according to the first test case that described acquisition module obtains, statement in described dynamic DBB has identical coverage information,

Second generation module, for concentrating the coverage information of each test case according to current 3rd test case, generate the current largest packet set of described dynamic DBB, largest packet in described current largest packet set comprises one or more described dynamic DBB, and each largest packet in described current largest packet set is concentrated identical test case to cover by described current 3rd test case;

Then add module, comprise: selection unit, for concentrating the test case selecting the grouping in current group set to be divided into the first grouping and the second grouping from current second test case, the described test case selected has different coverage informations to described first grouping and described second grouping, grouping in described current group set comprises one or more described dynamic DBB, and the initial value of grouping set is the largest packet set generated by the initial value of described 3rd test use cases;

Adding device, for obtaining the execution result of the test case that described selection unit has been selected, and adds described current 3rd test case to and concentrates by the described test case selected;

Judging unit, when all test case can not be concentrated to divide by described current second test case for each grouping in described current group set, judge that whether described current group set is identical with the largest packet set generated by described current 3rd test use cases;

Then described adding device, time also for being different in the judged result of described judging unit, grouping in described current group set is carried out merging to form largest packet, thus the coverage information of each test case is concentrated according to described first test case, continue to perform the operation that the described test case of part described second test case concentrated adds the 3rd test use cases to, wherein, being grouped in described current group set by grouping that the failure testing use-case of the correct test case of equal number and equal number covers of described merging; Or,

Described adding device, also for when the judged result of described judging unit is identical, completes the interpolation operation of described 3rd test use cases.

7. test cases selection device according to claim 6, it is characterized in that, described selection unit, specifically for concentrating the test case selecting described grouping to be divided into uniformly described first grouping and described second grouping from described current second test case.

8. the test cases selection device according to claim 6 or 7, is characterized in that, described selection unit concentrates the test case of selection to meet the following conditions from described current second test case:

$\arg \underset{t\∈T_g}{\max }\{\underset{i=1}{\overset{\left|G\left(T_r\right)\right|}{\mathrm{\Σ}}}S\left(g_i\right)*\mathrm{Split}(t,g_i)\},$ Wherein, Split (t, g _i)=min (| g _i1|, | g _i2|)

Wherein, S (g _i) represent the grouping g covered by the described test case t selected _iin comprise the degree of mistake statement, | G (T _r) | represent according to described current 3rd test use cases T _rthe quantity of the grouping formed, Split (t, g _i) the test case t that selected described in expression is by described grouping g _ibe divided into described first grouping g _i1with described second grouping g _i2in the quantity of a less dynamic DBB comprised.